The 5 Why failure analysis method is very popular in industry because of its simplicity. The trouble is the 5 Why method is useful only if you have real evidence of the true failure cause.
When you apply the 5 Why method to do root cause analysis of a failure it is very easy to give an answer to each 5 Why question asked. But how can you know that the answer to the 5 Why question is on the true failure path just because an answer is given?
A professional engineer I knew fell into the 5 Why trap. It is easy to go badly wrong once you start using the 5 Why root cause analysis methodology.
He got a team of maintainers and operators together to do a Five Why analysis for the root cause of a major equipment failure event. Standing at the whiteboard in front of the group he started by asking why did the failure event happen. A maintainer spoke up and the engineer wrote the answer on the board. From the answer he structured the next why question.
This time an operator answered the question and the engineer wrote it on the board. The third 5 Why question was then asked and answered, and so on it went until the 5 Why table was completed. The 5 Why root cause analysis took about 20 minutes to complete and a decision was made by the engineer to address the root cause. But the maintainers knew that the solution proposed would not fix the problem—the 5 Why analysis had not uncovered the true cause of the failure. This is the 5 Why trap! And the engineer walked right into it.
An answer to a 5 Why question may not be the true cause of an event. Look at the image below of a 5 Why cause and effect tree. Notice how there are five ‘or’ branches below the second event level. For the third level events many have multiple ‘or’ branches down to fourth level events. Any branch is an answer to a 5 Why question. But which branch is the right answer at each level of the 5 Why cause and effect tree?
The engineer took the first answer given as the cause of each 5 Why tree event level and he was soon deep down a hole that led to the wrong conclusions.
The 5 Why trap is to accept what someone tells you just because it is a believable answer to the 5 Why question.
There is only one sure way to protect yourself from the 5 Why trap: you must have evidence at each level of the 5 Why tree to confirm the branch that the failure events followed. If you have no evidence of the failure path you may end up fixing the wrong problem and the real failure cause left behind will repeat again.
Many people say that any 5 Why solution identified is still worth implementing because it will solve at least one cause of the 5 Why event. It sounds like a logical argument to make. But the real problem cause has not been solved and so it will surely reoccur. By doing work that does not stop the real cause you have wasted a lot of peoples’ time and spent money unnecessarily; and you have still left the real problem cause behind.
How to Successfully Use the 5 Why Method
To reduce your chance of falling into the 5 Why trap always start by developing the 5 Why event tree. Use the 5 Why analysis team to create an adhesive note cause and effect tree like in the above photo. By doing that you do not assume any answer is right. When the time comes to select the right branch use the evidence gathered from the event to confirm the path (if it is an ‘and’ event there will be multiple branches that you will need to follow).
If there is no evidence that proves the true 5 Why failure path, then use the ‘unexplained branches’ on the failure analysis cause and effect tree to identify what evidence to send the team away to investigate.
Once all the evidence is available use it to test and prove the branches of the real failure path. Should there be uncertainty as to which branch to take, use the team consensus response to the question. You want the team to debate uncertain answers so that you draw-on the team’s best knowledge and understanding of the situation. This puts the ‘wisdom of crowds’ into service for you. Later check if the branches further down the tree from the consensus answer are confirmed by the available evidence. If the evidence does not support the conclusion, or there is no evidence available, then the 5 Why team has lost its way.
REMEMBER, as soon as you find yourself ‘running blind’ in a 5 Why analysis and the evidence is not available to support the cause and effect path; you had better stop and backtrack before you fall into the deep hole of a 5 Why trap.
My best regards to you,
Mike Sondalini
Managing Director
Lifetime Reliability Solutions HQ